Comments on: JavaScript Performance: for vs. while

By: Kenneth

Kenneth — Sat, 31 Oct 2015 21:14:44 +0000

This is really mind boggling. I had always believe “While” is the better OPTION but these test proves a probability issue(What’s the chance that the time you used the loop will be the best). Based on this result the best average performance is “While”.

However, The number of times “For” performed better is more than the number of times “While” did, especially with that significant “While” spike only occurred once out of ten attempts.

Now that tells me there is a greater chance of having a faster result with “For” than “While”. So while “while” is better on average, I’m now led to believe “For” is the better option.

By: Tushar Singh

Tushar Singh — Thu, 28 Mar 2013 04:24:17 +0000

I ran 10 tests with 1 billion iterations and the average came out to be this
while: 14566.ms
for: 16297ms
(rounded figures)

so yea, while really is faster..

P.S change the font on textarea and inputs

By: Luca

Luca — Sat, 09 Jun 2012 20:48:52 +0000

Can you please try to replace i++ with ++i?

By: Giovanni

Giovanni — Fri, 03 Feb 2012 22:31:43 +0000

First, I have to say that I find the results interesting. Thanks for the research.
That beeing said, I agree with Derrick. Moreover, I think that this should not even be an aspect considered in the optimizazion of the code. I think that there’s always space for algorithmic optimization dealing with computational complexity, first (something much more important to optimize).
Finally, nothing guarantees that this “optimization” will work on every interpreter or on different versions of the same interpreter.
Interesting, nevertheless.

By: Derrick

Derrick — Wed, 25 Jan 2012 00:14:04 +0000

The ‘extremely’ minor performance difference isn’t worth the uglier code.

By: Aleksejs

Aleksejs — Tue, 24 Jan 2012 14:42:45 +0000

I think, if you have so diverse results (like 3 times slower “while” in second test), you have some process that disturbs normal testing. Anyway, you need to run not ten but ten million tests to be able to tell an average value properly.